Alternative Research Methods in Cognitive Psychology

1
Alternative Research Methods in Cognitive
Psychology

• Individual Difference Measures

2
Comprehension of 3D Anatomical Structures
3
(No Transcript)
4
Arrow View
5
Spatial Ability Measures

• Mental Rotation Test (Vandenberg Kuse, 1978)
• Guays Visualization of Views (Eliot Smith,
  1983)
• Scores were aggregated

6

Low spatial
Slice 10
High spatial
7
Coding of Drawings
Number of ducts Outside shape Spatial
relations between the ducts Spatial relation
between duct and outside shape
8
Study 1 Interactive vs. Non-Interactive
Animations

• Interactive group (N30)
• Allowed to rotate and pause the visualization at
  will during the task
• Motion-based depth cues
• Can access arrow view at will
• Non-interactive group (N30)
• Presented with a continuous looping rotation
  during the task
• Motion-based depth cues

9
Interactive group (N30) Allowed to rotate and
pause the visualization at will during the
task Non-interactive group (N30) Presented with
a continuous looping rotation during the task
10
Interpretation

• Interactivity seems to help performance
• What was it that helped the interactive subjects
  do the problem-solving?
• Control per se OR what they got to see?
• These groups did not see the same thing
• Interactive and non-interactive subjects received
  different visual information
• Need to match the visual input

11
Study 2 Yoked-Design

• Remove confound
• Examine the effects of interactivity
• Ensure active and passive viewers see same visual
  information

• Predictions
• Interactivity will confer an advantage
• Active participants will do better than passive
• Performance will be related to spatial ability

12
Method

• Between-subjects design N 60
• Spatial abilities measured
• MRT, Guay Test
• Randomly allocated to
• Active condition
• Passive condition
• Yoked pairs of participants

• Active participants controlled computer model via
  InterSense Inertiacube2
• Movements were recorded
• Later played back to a passive participant
• Visual information identical

13
Intertia Cube

• 3 DOF inertial orientation tracking system by
  InterSense

14
(No Transcript)
15
(No Transcript)
16
Results
No advantage to active participants
Relationship between performance and spatial
ability attenuated
17
Discussion

• No benefit from active control
• Finding was replicated
• How can we account for all our results?
• Does interactivity really help?

• Maybe the key variable is not interactivity per
  se, but what the viewer sees
• Active control allows the viewer to select
  relevant view for solving the problem

18
Access of Arrow View
19
Access to Arrow View affects Performance
Spatial ability and access to arrow view make
independent contributions to predictions of
performance. Together account for 39 of
variance.
20
Study 3 Arrow Views vs Interactive

• Condition 1 Interactive (intertia cube
  interface)
• Condition 2 Passive Viewing
• Starts at page view
• Rotates to arrow view
• Jitters at arrow view

21
Arrow Views group performance better than
Interactive group
22
Access of Arrow View
23
Difference due to Low Access Group
24
Summary (Cross-Section Task)

• Highly correlated with spatial ability
• Assisted by an interactive animation, however
• Not all individuals access the most informative
  view
• Performance also related to access of the most
  informative view

25
Summary (Methods)

• Correlational
• Drawings to study mental representations
• Interactivity logs another rich source of data
• Individual differences add another dimension

26
Applications for your future

• User Experience Research
• Local company in UCSB Job Bank (Call Wave) We
  are looking for a talented new member of our team
  to provide us with quantitative and qualitative
  data on users' experience with our products.
  Plan, design and conduct user research including
  field studies, user observations, surveys and
  focus groups. Synthesize and analyze data on user
  behavior and preferences to identify patterns and
  trends in utilization.
• The successful candidate will have a BA/BS degree
  in Ethnography, Anthropology, Human Factors,
  Human-Computer Interaction, Cognitive or
  Experimental Psychology, Cognitive Science, or
  related field.
• Youve learned all the rigorous experimental
  methods industry does a slapdash job of applying
  these but its better than nothing. Go onward and
  change the world!

27
Questions?

• khooshabeh_at_psych.ucsb.edu

28
Extra Slides
29
Informal Task Analysis
encode stimulus object
assume arrow perspective
rotate animation
mental perspective taking
mental rotation
imagine part of object cut away
imagine resulting cross section
draw cross section
30
Cross Section Task
Central to Learning anatomy Interpreting
medical images
31
Analyzing Patterns of Interactivity
32
Method of Analysis

• Python simulation program generates a log file
  with time-stamped (x, y, z) position data
• MATLAB used to generate histogram frequency plots

33
Interactivity Plots

• Plots arranged as pairs in a columnar
• Top, middle, and bottom pairs show (y, z, x)
  (yaw, roll, pitch)
• Upper plot in the pair has angle on the abscissa
  and indicates number of time-bins spent at each
  location
• Lower plot in each pair has time on the abscissa
  and indicates changes over time

34
Start position
Test Page view
Most informative view (arrow view)
35
Yaw
Roll
Pitch
36
(No Transcript)
37
(No Transcript)
38
(No Transcript)
39
(No Transcript)
40
(No Transcript)
41
(No Transcript)
42
(No Transcript)
43
General ConclusionsApplications for Selection

• Performance in a variety of medical tasks is
  correlated with spatial ability
• Applications for selection
• Low spatials can acquire skills.
• With training effects of spatial ability are
  attenuated,
• But still there!
• Ability-skill distinction may be artificial

44
Applications for Training

• Promising role for new technologies
• Simulations can be an effective means of
  training, minimizing risk to patients
• External visualizations can augment internal
  visualization
• However individual differences in use of
  visualizations simulations too
• Importance of metacognitive skills

45
Current Directions

• Stereoscopic displays (Petes Minicon Talk)
• Training in internal visualization of cross
  sections (Cheryls dissertation)
• More effective external visualizations to
  minimize disorientation of low-spatial
  individuals (with Andy Stull)